Pydantic returns 'field required (type=value_error.missing)' on an Optional field with a custom model

Question

I'm trying to build a custom field in Fastapi-users pydantic schema as follows:

class UserRead(schemas.BaseUser[uuid.UUID]):
    twitter_account: Optional['TwitterAccount']

On UserRead validation Pydantic returns

field required (type=value_error.missing)

on every field in 'TwitterAccount' schema.update_forward_refs() is called at the end.

TwitterAccount itself has required fields and making them optional isn't an acceptable workaround. I notices I could make Optional[List['TwitterAccount']] and it will work, but that's a bit silly.

Answer 1

Optional is a bit misleading here. What it means technically means is that twitter_account can be a TwitterAccount or None, but it is still a required argument. To make it truly optional (as in, it doesn't have to be provided), you must provide a default:

class UserRead(schemas.BaseUser[uuid.UUID]):
    twitter_account: Optional['TwitterAccount'] = None